2.0 Analysis 2.1 Vessel Operation in Pilotage Waters Problems associated with shipboard practices in pilotage waters continue to be of concern in the maritime industry. The IMO has put forward measures to improve training that, over a period of time, will lead to higher standards and greater safety. Meanwhile, occurrences involving vessels operating in pilotage waters continue to draw the attention of the Board. The analysis of what happened in this occurrence, and why, is presented within the framework of BRM-related issues, and fatigue-related issues. 2.2 BRM-related Issues 2.2.1 Marine Culture and Team Work Navigation with a pilot on board creates a situation where a relative stranger, the pilot, is teamed with an existing crew to carry out a coordinated job together. While the pilot has the local navigational knowledge to analyse local cues more readily and take rapid action as necessary, the ship's crew has a greater understanding of the ship's handling characteristics. Because pilots, masters and officers of vessels have different areas of experience and training, it is essential that the skills of each be combined in the working relationship of a bridge team. Pilots have to work with ships' complements whose competency, proficiency, language skills and workload vary substantially from vessel to vessel. This has resulted in some pilots generalizing that they cannot rely on the ship's complement to assist in navigation. While these issues are being addressed through the introduction of STCW and ISM, the reluctance to utilize the ship's complement has become part of the culture of the pilot.[21] Furthermore, in many cases a ship's complement will adopt a passive approach to the navigation of a vessel under pilotage. The result is that some pilots may feel it necessary to take on an additional workload which, in certain circumstances, may be beyond their ability to handle. This is contrary to the principles of BRM and the safe operation of a vessel. In this instance, the fact that the OOW was new to both the vessel and the area did not diminish his ability to plot the vessel's position, monitor her progress and meaningfully participate in her safe navigation. The OOW, by querying the pilot and plotting the vessel's position at regular intervals, demonstrated that he could effectively participate in the navigation of the vessel. Despite this, the pilot did not take full advantage of the potential for the OOW to assist in the navigation of the vessel. The master was not aware that the pilot's workload was increased due to traffic, nor that he had elected not to enlist the services of the other pilot or to effectively use the OOW in the essential tasks of position fixing and/or collision avoidance. The pilot's decision to take on the additional workload may have been based on generalized assumptions regarding the OOW's capabilities, rather than on an informed assessment. This decision left the pilot relying solely on his own performance, foregoing the potential for team work, and leaving little room for error. 2.2.2 Situational Awareness Similarity of Forward Bay, Boat Bay and Baron Reef-Growler Cove Area Proceeding westbound along the south shore of West Cracroft Island and before reaching the Baron Reef-Growler Cove area where the course alteration into Blackney Passage is effected, a vessel passes Forward Bay and Boat Bay. A review of the chart for the south shore of West Cracroft Island reveals similar shoreline indentations in the Forward Bay, Boat Bay and Baron Reef-Growler Cove areas (6M and 4 M apart, respectively). To an observer on a vessel proceeding eastward or westward along the shore, a cursory radar observation could result in improper identification of the three areas. Pilot's Loss of Situational Awareness At 0113 the pilot reported being off Boat Bay Light, located on the western headland of Boat Bay (Swaine Point), when the vessel was in fact passing the western headland of Forward Bay. When the vessel was approaching Boat Bay, the pilot initiated the course alteration which was to have taken place off Baron Reef. The grounding position reported by the pilot to MCTS was in the vicinity of Blackney Passage, which is consistent with a mental model in which the vessel's course had been altered to enter Blackney Passage. This suggests that although the pilot was viewing the radar, his interpretation of the radar presentation was at variance with the actual radar display and that he did not make effective use of the radar. Between 0100 and 0130 the pilot's workload increased as he was engaged in collision avoidance manoeuvres with southbound vessels. To effectively monitor traffic for collision avoidance, the radar used by the pilot was switched to shorter ranges. The use of shorter ranges hindered continuous monitoring of the shoreline, and the increased workload associated with collision avoidance may have interfered with navigation. In dealing with the cumulative effect of the increased workload, the pilot was apparently unable to process all the navigational cues and information available to him. As a result, the pilot transposed his position from Forward Bay to off Boat Bay Light. There were a number of cues that could have alerted, but did not alert, either the pilot or the OOW to the navigation error. These included: the unique characteristics that differentiated Forward Bay and Boat Bay. These include the absence of a navigational light on the headland west of Forward Bay; however, given the deteriorating visibility, an assumption was made that the light was obscured by fog; the mismatch between the 0113 position report to MCTS and the OOW's 0115 plotted position. The OOW did not detect the pilot's error in reporting the 0113 position to MCTS; the vessel would have had to average a speed of some 19 kn to have been off Boat Bay Light at 0113 instead of 0135. This is an unrealistic speed in the given circumstances. Neither the pilot nor the OOW used the vessel's estimated steaming time between course alteration points as a means to cross-check whether the vessel had reached the alter-course position. OOW and Situational Awareness The OOW was new to the area and did not possess the necessary local knowledge; the only member of the bridge team who had local experience of the area was the pilot. Further, the pilot did not specifically indicate to the OOW that the vessel's course was being altered to enter Blackney Passage. In the absence of complete information, the OOW could not be expected to have, nor did he have full situational awareness and he was only generally following the vessel's progress. Because the pilot had not passed all relevant information to the OOW, the latter's mental model had been developed on fragmentary information. Consequently, it was readily influenced after the pilot indicated to the OOW that the vessel had reached the alter- course position. The subsequent rapid progression of events precluded the OOW from plotting the vessel's position and upgrading his mental model--which is essential in a course alteration situation--as by now the OOW's emphasis had shifted to ensure that the mechanics of the course alteration were properly executed. It is likely that incorporating new information--which became available from the pilot--was made more difficult by the increased workload of the rapidly developing situation. 2.2.3 Limitations Imposed by Navigational Practices This occurrence highlights the need for appropriate distribution of the workload within the limits of individual expertise and capacity. Pilotage is mentally demanding and, consequently, pilots often implement tools and practices to facilitate pilotage operations. The pilot's course book is an example of such a tool. Relying solely on his course book as a passage plan, and without the use of an additional aide-mmoire, the pilot became confused as to which leg of the passage the vessel was on. It was his belief that the vessel was approaching the course-alteration point in Blackney Passage after having inaccurately reported passing Boat Bay Light some 20 minutes prior to the time he would have passed it. His belief could have been influenced, in part, by the fact that: having made several course alterations to starboard for traffic, the last heading prior to course alteration was 285, close to the 290 heading which the course book indicated was the intended course to steer to arrive at Blackney Passage; the content of his radio communication with the ISLAND PACER both in terms of that vessel's geographical location and the weather being experienced by her did not contradict his mental model that the RAVEN ARROW was approaching Blackney Passage, but instead could have reinforced this belief. Once a hypothesis or a certain way of thinking about a problem is adopted, it is very resistant to change. Consequently, one becomes susceptible to what has been termed confirmation bias or hypothesis locking, where one may selectively attend to information that confirms one's mental model, and disregard other--equally valid--information that would disprove one's hypothesis.[22] The effect of hypothesis locking is so great that it may take the intervention of another person with contradictory information to overcome it. 2.2.4 Communication between Pilot and OOW It is possible that when the pilot began the course alteration to 320, under the mistaken impression that the vessel had reached Blackney Passage, the OOW may initially have been under the impression that the pilot intended to alter course to 290(having reached the Boat Bay Light). However, it would appear that the OOW had some doubt with respect to course alteration and he questioned the pilot as to whether the vessel had reached the alteration point. The OOW readily acquiesced when the pilot responded in the affirmative. Because communication between the pilot and the OOW was minimal and imprecise and because the pilot had not used the services of the OOW to advantage in the navigation of the vessel, a barrier likely had been created that left the OOW less likely to challenge the pilot's decision, especially at a critical time of course alteration. This could have been reinforced by the tendency of OOWs generally not to distract or interfere with pilots. On the other hand, effective monitoring of the vessel's progress would have heightened the OOW's situational awareness such that he could have sought clarification from the pilot with regard to the course alteration. However, this was not done. The pilot's specific plan was not explained to the OOW, which precluded the latter from effectively participating in the navigation. Although the OOW plotted the vessel's position at 15-minute intervals, this information was not used by the pilot. On the other hand, the OOW did not effectively monitor the pilot's communication with MCTS, in that he was not alerted to the pilot's error in reporting the 0113 position. Thus, the first opportunity to take remedial measures, some 20 minutes prior to the grounding, was lost. Subsequently, when the OOW became aware of the pilot's intention to alter course, he took the bearing and distance from the radar to establish the vessel's position. However, because the pilot commenced the course alteration before the vessel's position could be plotted on the chart, another opportunity to verify the position and remedy the error was lost as the focus now shifted to prompt execution of the pilot's orders. For the OOW to carry out his responsibilities as the master's representative, it is essential that all communication with MCTS be monitored, be it merely position reporting or traffic-related. Further, improper monitoring of MCTS communication has the potential to jeopardize the safe navigation of the vessel. In this instance, the OOW only casually monitored the pilot's communication with the MCTS. This would suggest that the necessity to effectively monitor the pilot's communication with the MCTS was not fully appreciated by the OOW. The use of sound navigational practices to monitor the vessel's progress, i.e., communication and sharing of information, and the use of more than one method of ascertaining the vessel's position, would have better enabled both the pilot and the OOW to realize the error in ample time to initiate remedial measures. 2.2.5 BRM, Pilotage and Safety In an era of rapid change, with increasingly large ships and decreasing crew sizes, and commercial pressures becoming part of everyday operations, the services provided by the pilots should reflect this reality. As the pilot has the conduct of the vessel and possesses local knowledge, and as complete support of other members of the bridge team, especially the OOW, is essential for the safe navigation of the vessel, good seamanship practices dictate that a passage plan--approved by the master and the pilot, and containing all pertinent information--be suitably displayed at the conning position or radar display. The absence of such a plan prevents the ship's personnel from effectively monitoring the vessel's progress, and results in inopportune communication--to seek clarification--that has the potential to interfere with pilotage at a critical time in a manoeuvre. In this instance there was no agreed-upon passage plan aboard the RAVEN ARROW, and the courses as laid on the ship's chart differed from the pilot's approach to rounding Cracroft Point. Although the pilot had taken BRM training, on this passage he did not put the associated techniques into practice. Neither the pilotage authorities nor the BCCP have a regime in place to monitor how BRM principles are being implemented in practice. The absence of such a monitoring regime permits the old culture to continue, thereby effectively negating the benefits of such training (to the detriment of safety), as was the case aboard the RAVEN ARROW. This occurrence reinforces the need for implementation of TSB recommendation No. M95-08, which called for: an agreed-upon (approved) passage plan prior to the commencement of passage in pilotage waters, and for the pilot to encourage a climate whereby bridge team members can comfortably provide input. While the pilot has the conduct of the vessel, the master retains command of the vessel. In order to retain effective command of the vessel (by the master or, in his absence, the OOW), masters need to hold objective discussions with pilots who are less inclined to team work. This need has been widely recognized by a number of non-governmental maritime organizations including the International Association of Independent Tanker Owners (INTERTANKO) and the International Chamber of Shipping. While the intent of the aforementioned TSB recommendation had been accepted by TC, and discussions have taken place between TC officials and pilotage authorities, concrete agreed-upon measures have not yet been established. 2.3 Emergency Action At the time of the course alteration, the vessel was four cables off shore. In a partly loaded condition, the vessel's turning circle is 3.3 cables. The course recorder trace shows that the vessel's swing was arrested on a heading of about 320 for a short period before the pilot realized his error. The pilot, in order to extricate the vessel from this situation, had the option to either come around over 120 degrees to starboard or over 60 degrees to port to clear the area. The pilot considered that the vessel was still four cables off shore and ordered hard-a-starboard. However, the distance off land was less than the turning circle of the vessel and the vessel grounded. Sea room was available to go to port with a greater chance of extricating the vessel, though the alteration of course to port would have required some adjustment for traffic. The pilot's assessment that the vessel was still four cables off shore is consistent with his thinking that the vessel was in Blackney Passage where such a manoeuvre might have been possible. The grounding position initially reported to MCTS by the pilot was in Blackney Passage. As part of the attempted clearing manoeuvre, the engine could have been put to emergency full astern, but the pilot was reluctant to use the engine because it was on a 30-minute standby. Although the main engine was available, it was not used. 2.4 Pilot Scheduling and Safety 2.4.1 Fatigue and the Pilot's Performance Fatigue is a physiological state characterized by impaired performance and diminished alertness. Two causes of fatigue are inadequate quantity or quality of sleep, and disruption of circadian rhythm. These can result from irregular work schedules, extended duty, or altered work/rest schedules, and have been identified as contributory factors in many industrial accidents.[23] Research suggests that it is not possible to store sleep. As a person remains awake, a sleep need develops, notwithstanding how well rested the individual was at the beginning of the wake cycle. Most people need between 7.5 to 8.5hours of sleep per day. A person who does not obtain required sleep will develop a sleep debt and will be subject to performance degradation. A rough calculation of sleep debt can be determined by allowing two hours of credit for every hour of sleep up to a maximum of 16 hours credit and one hour of debit for every hour awake. Figure 4 shows pilot No.1's sleep history and accumulated sleep debt prior to the occurrence. Work shifts are represented by a thick black line, while time off duty is represented by a thin black line. Line A on the graph represents the pilot's sleep credits/deficits based on the recorded information as provided by the pilot. From this diagram, it is evident that the pilot had accumulated a sleep debt equivalent to having missed approximately 10 hours of sleep--or just over one full night's sleep--over the three days preceding the accident. At the time of the grounding the pilot had been awake for 19.5 hours and would have been awake for 21 hours upon completion of the planned watch. The need for sleep typically recurs after about 15 or1 6 hours of being awake, even for someone who is well rested. On September 21 the pilot's assignment aboard the MERLION AC lasted for 13.5 to 14 hours, of which some 3 hours were spent on pilotage duties in Canadian waters. After allowing for dispatch, transportation, vessel boarding and formalities, it is reported that the pilot could have had an opportunity to obtain sleep for some four hours on this assignment. The impact of such sleep is depicted by line B, and shows that his sleep debt at the time of the occurrence would have been drastically reduced. This sleep debt could then have been further mitigated by the pilot having obtained sleep between 1800 and 2300 on September 23 while off-duty, as shown by line C. Consequently, the pilot's schedule over a short period of six days was not a factor in this occurrence. In today's workplace, the loss of performance associated with alcohol is not tolerated, while that associated with fatigue--with similar results--is tolerated. Researchers at the Defence and Civil Institute of Environmental Medicine found that after 18 hours awake, people showed a 30 per cent decrement in performance on cognitive and vigilance tasks, and after 48 hours the impairment averaged 60 per cent. Researchers at the Centre for Sleep Research at the University of South Australia found that after 18 hours without sleep, students performed as poorly on performance tests as they had with a blood alcohol concentration of 0.05 per cent. After 24 hours without sleep, their performance dropped to that of a person with a blood alcohol concentration of 0.096 per cent.[24] That pilot No.1 may have been fatigued is corroborated by the observation of the second pilot, that the demeanor of pilot No.1 was suggestive of his being tired. It is likely that pilotNo.1 underestimated his tiredness, as people are poor judges of their own levels of fatigue and alertness. It has been demonstrated that individuals (especially sleepy individuals) do not reliably estimate their alertness and performance.[25] Fatigue can lead to forgetting or ignoring normal checks or procedures, reversion to old habits, and inaccurate recall of operational events. Fatigue can also reduce attention, the effects of which are that people overlook or misplace sequential task elements, become preoccupied with a single task, and are less vigilant. When alertness is impaired, people may fix their focus on a minor problem (even when there is risk of a major one), they may fail to anticipate danger and they may display automatic behaviour syndrome. Problem solving can also be affected and flawed logic may culminate in the application of inappropriate actions. In this occurrence, pilot No. 1 appears to have demonstrated several errors involving reduced attention and memory typical of the fatigued state: the pilot jumped a line in the course book following the course alterations taken to avoid fishing vessels; the pilot transposed his position from Boat Bay to Blackney Passage subsequent to receiving radio traffic from another vessel which was entering Blackney Passage; the pilot interpreted the radar image to be that of a position further along his route; the pilot became preoccupied with altering course to enter Blackney Passage, and only made use of confirmatory information; the pilot altered course to starboard although there was ample sea room available to swing the vessel around to port and make a fresh approach (overshooting the alter-course position to enter Blackney Passage would not have placed the vessel in danger of grounding); and the pilot was unable to alter his erroneous mental model even after the vessel grounded, as is evidenced by his reporting the vessel's position incorrectly. While fatigue is not necessary to explain these forms of error, they are consistent with those expected to be made by a fatigued individual. Hence it is probable that fatigue was a contributory factor in this occurrence. A review of the OOW's work/rest schedule revealed that September 22 was a day in which he had a relatively full workload. Despite this, his fatigue is not considered to have been a factor in this occurrence. 2.4.2 Pilot Scheduling Issues Acknowledging that operational considerations will influence scheduling, a preliminary review of the pilotage schedule suggests that the following factors merit further consideration: irregularity of the pilot's schedule; special conditions affecting midnight shift workers; work/rest cycles; and off-duty recovery periods. 2.4.2.1 Irregularity of the Pilot's Schedule Figure 5 - Pilot's schedule, September 19-24 Figure 5 presents pilot No.1's schedule for the period September 19 to September 24. The irregularity of the pilot's schedule is evidenced by the fact that the pilot had worked three midnight shifts in six days, and that interspersed with these midnight shifts were an evening shift and a day shift. Over the six days, off-duty periods between the shifts vary in length from about 5 hours to about 21 hours. Studies of marine watchkeepers who work odd shifts have shown that shorter/poor-quality sleep and poorer performance are associated with irregular work/rest cycles.[26] 2.4.2.2 Special Conditions Affecting Midnight Shift Workers Even small reductions in sleep--of as little as two hours--can result in measurable changes in performance on tests of vigilance. Many studies of shift workers link sleep loss with the time of day that sleep is taken.[27] Sleep taken during the day is shorter and of poorer quality than sleep taken at night. Studies have shown that pilots who conduct pilotage around midnight produce high levels of adrenaline (up to seven times that of a normal person, sleeping) and it can take as long as two days for adrenaline levels to return to normal.[28] For such a pilot, sleep will be very difficult. Hence, working at night cannot be taken as equivalent to working during the day. Sleep is poor if it is taken at times of day when the body is physiologically alert, and sleep loss exacerbates the normal drop in alertness and performance found at circadian low points in the day.[29] Statistics for 1996 show that assignments with a pilot order time between 2200 and 0600 numbered some 4,650 assignments and accounted for 38 per cent of total assignments (Figure 6). Figure 6 - Assignments by time of day ordered 2.4.2.3 Work/Rest Cycles A pilot may work for many consecutive days, at various times of day, and for as long as eight hours per shift if working alone (and up to six hours per shift if working in conjunction with another pilot). A pilot might work for up to eight hours, be off duty for eighthours plus travel time, and then return to work at the end of this period (multiple shorter shifts may have abbreviated off-duty time associated with them). Moreover, such a pattern could continue for many days, so that the individual's work/rest schedule would never develop a consistent circadian pattern. 2.4.2.4 Off-duty Recovery Periods Recovery from an acute sleep deficit, cumulative sleep debt, prolonged performance requirement, or extended hours of continuous wakefulness is another important consideration. Operational requirements can engender each of these factors and it is important that a recovery period provide an opportunity to acquire recovery sleep and to re-establish normal levels of performance and alertness. Frequent recovery periods are important. More-frequent recovery periods reduce cumulative fatigue more effectively than less-frequent ones. For example, weekly recovery periods afford a higher likelihood of relieving acute and cumulative fatigue than monthly or yearly recovery periods.[30] The current schedule adopted by the BCCP makes use of monthly recovery periods, through the use of a 20-days-on and 10-days-off cycle followed by an extended leave period of 30 or 50 days off, at the end of the 80-day cycle. However, the opportunity to rest and recuperate during the 10-days-off period is diminished by call-backs. The adverse effects of the schedule are reflected in some pilots' complaints to the BCCP that they tend to become fatigued towards the end of their 80-day cycle during the peak season. 2.4.3 Increased Demand for Pilotage Services, Pilot Fatigue and Effects on Schedule Over the last few years there has been a steady increase in the demand for pilotage services by ships calling at west coast ports or transiting coastal pilotage waters. The increase is primarily associated with the growth of cruise ship operations--one of the growing industries on the west coast of Canada--and most of the extra assignments are long trips requiring two pilots. The current scheduling practices were set up before this recent boom. Despite the employment of ten seasonal pilots and two half-time pilots to supplement the full-time pilots, the workload is substantially higher in the peak season than in the rest of the year. The 1996 statistics show that the increase in pilotage assignments varied by some 35 per cent between the months of January and August, but the increase was some 56 per cent when the least-busy and the most-busy thirty-day periods are compared.[31] Thirty-day period figures have been used in this report. Information provided by BCCP indicates that during the peak season pilots are kept very busy during their 20 days on duty, and that the pilots' average 20-day workload of 8 to 10 assignments in the off season rose to 14 to 21 assignments in the peak season. Although the greater part of the increase involved longer trips, to which two pilots are allotted, there is still a significant increase in work hours per assignment for each pilot, as the long trips range to over 27 hours while the short trips are typically a couple of hours in length. With pilots spending longer periods away from their home base on these long assignments, there are fewer pilots available to carry out other assignments that arise during their absence. The situation is further aggravated by the short rest periods associated with the watch system the pilots must adopt on long trips. Pilotage operation requires high levels of concentration which, when combined with a heavy workload, can result in high levels of fatigue.[32] Some significant factors that adversely affect pilot performance and contribute to pilot fatigue include the following: The inherently irregular work/rest schedule associated with pilotage duties, which prevents a pilot from developing a consistent sleep pattern. That 38 per cent of BCCP assignments are carried out between 2200 and 0600, where rest must be in the form of daytime sleep. The quality of sleep taken during the day is poor, compared to that taken during the night, but time off between assignments is not adjusted to reflect this. Off-duty recovery periods are linked to a 30-day cycle, whereas more-frequent recovery periods would reduce the chance of cumulative fatigue. The opportunity to rest and recuperate during the pilot's 10-days-off period is diminished by call-backs, a situation exacerbated in the peak season. This annual increase in demand on the pilots' services is not recognized by any adjustment in the overall pilots' work schedule. No regime has been instituted by which more pilots would be on duty during the peak season, either through an adjustment to the 30-day cycle or an adjustment to the number of pilots on extended leave. It would appear that there is a tendency for a pilot following the present schedule to suffer from cumulative fatigue towards the end of the 80-day cycle during the peak season. This is substantiated by complaints to this effect that the BCCP has received from pilots. There are a large number of pilot call-backs to meet the additional demands placed on pilotage services in the peak season and for some of these the pilot gets as little as two to three hours' notice. Every effort is made to recall, first, those pilots who are on their extended leave periods, but if none is available (as is often the case), pilots on their 10-days-off period are called. The onus is on the individual pilot either to accept or decline the call-back assignment. However, as pilots on extended leave are often unavailable, those on their 10-days-off period have little option but to accept these additional assignments; the pilot population as a whole is contractually obliged to handle these assignments. A review of current scheduling practices shows that an effort is being made by the PPA and the BCCP to reduce work-related fatigue over the 20-day duty period by introducing policies on mandatory rest periods between assignments and on the sequence in which pilots are recalled. These policies need to be expanded to meet longer-term requirements and address the factors leading to cumulative fatigue in the 80-day cycle, especially in the peak season. An overview of the existing structure suggests that the only way the current scheduling system can meet peak-season demands on the pilots is to increase the number of pilot call-backs, which are already at over 260 per year. Call-backs are unscheduled and disruptive to the pilots, and the scheduling system needs to be more flexible to meet periods of high demand. Although the yearly increase in the demand for pilotage services is somewhat predictable, prudence dictates that this increasing demand not be considered an ongoing phenomenon that must necessarily translate into an increase in staff; competition between ports has the potential to produce wide fluctuations in vessel calls at B.C. ports. Consequently, the fluctuating demands in annual pilotage services, especially those associated with peak periods, need to be met by a flexibility that does not exist in the present work schedule. The Board, concerned about the effects of fatigue on pilots' performance associated with pilot scheduling, recommended in TSB Report No. M93C0003 that the GLPA implement measures to address work-related fatigue.[33] Typical of measures introduced to address this issue is an amendment in the Collective Agreements that a pilot having worked two consecutive nights may ask to be not dispatched before 0600 the following morning.[34] 2.4.4 Awareness of the Effects of Fatigue and the Impact of Scheduling Practices There are many factors likely to affect the level of fatigue and alertness that a pilot will experience. The pilots can be subjected to two types of fatigue; one is work-related and the other is general fatigue brought about by their lifestyles. Neither the PPA, nor the BCCP, nor the pilot has full control over all of these factors. The PPA and the BCCP do not have control over what a pilot does during time off. Conversely, the pilot does not formulate the work schedule. Neither has control over delays between the pilot order time and a vessel's readiness to begin pilotage. Mitigation of fatigue, therefore, has to be a joint enterprise among all involved. Time off permits a pilot to take a break from work-related stress and to recuperate before the next call of duty. There is no regime in place, be it self-monitoring or otherwise, to ensure that the pilots are sufficiently rested to carry out their assignments. A fatigue management plan that sets out operating rules and procedures, with well-defined responsibilities, would greatly assist in addressing fatigue-related issues. In this instance, although pilot No. 1 had some 21 hours for rest and sleep between assignments, prior to boarding RAVEN ARROW, he showed signs of fatigue at the time of the occurrence as observed by pilot No. 2. While recognizing the symptoms of fatigue, pilot No. 2 did not fully appreciate the impact of such a state, highlighting the need for more formalized education and training to help pilots and management better appreciate the negative effects of shift work on performance. Knowledge of such factors can influence how pilots acknowledge and deal with problems of fatigue and alertness. 2.5 Quality Control and Pilotage The need to have clearly defined responsibilities in the safe management and operation of ships has been recognized by the IMO, and the International Safety Management(ISM) Code is intended to address this issue. As neither the owner, nor the agent, nor master has in-depth knowledge of the pilot's history of assignments--and sleep history--they have no way of knowing the minimum level of service the pilot would provide. Nonetheless, the master assumes full responsibility for the pilot's performance. As the prime purpose of pilotage is safety and, as compulsory pilotage areas are established for the benefit of the community to protect the environment and the port infrastructure from marine accidents, there is an expectation that the pilot's performance and operational procedures are of a standard that is internationally recognized and accepted.[35] Thus, in an era when shipowners have made a commitment to quality control, safe operation of a vessel in compulsory pilotage waters can be assured through an established, well-documented quality assurance program adopted by pilotage authorities, such as ISO 9002. This occurrence highlights the need for a safety management system that incorporates a fatigue management plan with clearly defined responsibilities. 3.0 Conclusions 3.1 Findings the pilot's sole reliance on the course book, without an additionalaide-mmoire, resulting in confusion as to which leg the vessel was on; the pilot losing track of time; and cursory radar observation, unduly influenced by traffic communication. the pilot's sole reliance on the course book, without an additionalaide-mmoire, resulting in confusion as to which leg the vessel was on; the pilot losing track of time; and cursory radar observation, unduly influenced by traffic communication. The cumulative effects of the workload associated with collision avoidance, and the changing of the range on the radar, resulted in the pilot not being alerted by some of the available cues. The OOW plotted the vessel's position at frequent intervals but the pilot elected not to utilize the information. Still under the impression that the vessel was in Blackney Passage, the pilot attempted corrective action to starboard when he realized that the vessel was not at the alter-course position. The OOW did not effectively monitor the pilot's communication with Marine Communications and Traffic Services (MCTS) and was not alerted when the pilot erroneously reported the vessel's position to MCTS some 20 minutes prior to the grounding. Bridge Resource Management in Pilotage Waters The pilot had received bridge resource management (BRM) training but he did not put into practice all relevant BRM elements. Neither the Pacific Pilotage Authority (PPA) nor the British Columbia Coast Pilots (BCCP) has a regime in place to monitor the application by the pilots of sound BRM practices. The pilot culture tends to lead some pilots to generalize that the ship's complement will provide minimal support in the navigation of the vessel, without assessing individuals on their merits. The ship's passage plan had not been agreed upon and this precluded the ship's personnel from effectively monitoring the vessel's progress. The pilot's lack of communication with the OOW created a climate on the bridge that made the OOW less likely to participate in the navigation of the vessel. The lack of effective communication between the pilot and the OOW initially resulted in the mental model of the OOW being at variance with that of the pilot. The OOW did not fully appreciate the need to effectively monitor the pilot's communication with the MCTS. As effective monitoring of the vessel's progress was not carried out by the OOW, he did not have full situational awareness and he did not seek clarification from the pilot with regard to course alteration. The OOW's mental model was based on fragmentary information and would account for the OOW being readily influenced by the pilot's actions/decisions. Fatigue in Pilotage Waters Time off between on-board watches and between assignments was not effectively used for sleep by pilot No. 1, and he was probably fatigued at the time of the occurrence. Pilot No. 1 may not have fully appreciated the negative effects that irregular work schedule and sleep debt can have on performance. There is no formalized education/training program in place for BCCP pilots and management that would provide guidance to pilots regarding conditions conducive to fatigue and the impact of scheduling on fatigue. Neither the PPA nor the BCCP have a regime in place to monitor pilot fatigue. Pilot Scheduling The demand for pilotage services increases about 56 per cent during the peak season, compared with the rest of the year. Despite the use of additional part-time/seasonal pilots to complement full-time pilots, there are a large number of call-backs to meet pilotage demand during the peak season. Call-backs, which are primarily in the peak season, increase workload. The current scheduling system lacks the flexibility necessary to deal with peak season pilotage demands, and pilots who follow the schedule suffer from fatigue towards the end of the 80-day cycle. The use of monthly recovery periods in the current scheduling system, instead of more-frequent recovery periods, reduces the chance that a pilot will fully recover from fatigue. 3.2 Causes The RAVEN ARROW grounded in fog when the pilot lost situational awareness and prematurely altered course to enter Blackney Passage after having elected to conduct the navigation of the vessel without assistance from the ship's complement. Contributing to the occurrence were the following factors: the pilot was probably fatigued; sound navigational principles were not implemented by the bridge team; the exchange of information between the pilot and the OOW was minimal and imprecise; and the OOW did not effectively monitor the pilot's communication with Marine Communications and Traffic Services. 4.0 Safety Action 4.1 Action Taken 4.1.1 BRM Training Requirement for Pilots Following the occurrence, all four Canadian pilotage authorities have recognized the need for BRM training and recommended that it be compulsory for their pilots. Further, Transport Canada is proposing amendments to the General Pilotage Regulations to include a requirement for pilots to be trained in BRM. 4.2 Action Required 4.2.1 Pilot Fatigue In the marine transportation industry, work schedules and working conditions are often conducive to degradation of performance due to fatigue. With respect to pilots, the irregular work/rest schedule, the high number of night assignments, the additional workload associated with seasonal fluctuations in demand for pilotage services, and diminished opportunity to rest/recuperate (as highlighted in this occurrence) all have the potential to adversely affect pilot performance. On the west coast, the BCCP determines the number of pilots available at a given time, while the PPA administers the scheduling of the pilots. During the seasonal peak periods, there is an increase in demand for pilotage services that places additional workload on pilots. The number of pilots on duty and their overall schedule are not adjusted to reflect the seasonal fluctuations in demand for their services. Although part-time and seasonal pilots are used, the heavy demand in peak periods still results in a large number of call-backs of off-duty pilots. This heavy demand causes some pilots to suffer from cumulative fatigue during the peak season. Indeed, pilots have made complaints to the BCCP to this effect. While the Board recognizes the efforts made by the PPA and the BCCP to address work-related fatigue over the short-term single on-off cycle period, it is concerned that the issue of chronic fatigue over the long term, which involves multiple work cycles between the pilot's extended leave periods during the peak season, remains unaddressed. This is not an isolated occurrence where fatigue has been recorded as a factor contributing to an accident involving a pilot. In a previous occurrence involving the striking of a bulk carrier and a Canadian tanker, the report attributed degradation in the pilot's performance to work-related fatigue due to insufficient rest (TSB Report No. M93C0003). In its report, the Board had expressed concern that the system places pilots in the difficult position of making a safe decision, while standing to gain financial benefit from accepting the extra assignment. Consequently, the Board recommended that TC and the Great Lakes Pilotage Authority (GLPA) implement policies and procedures for allocating pilotage assignments so as to minimize the adverse effects of fatigue on performance and provide training to pilots on managing fatigue.[36] It is understood that the GLPA now has formalized training on fatigue management for its pilots. In a 1997 occurrence, a Liberian bulk carrier went aground twice in the St. Lawrence River under the conduct of a pilot. (The investigation, M97L0030, is ongoing.) When the first grounding occurred, the pilot had had the conduct of the vessel continuously for over seven hours. More than 12 hours later, the vessel grounded a second time under the conduct of the same pilot. By the time of the second grounding, the pilot had been on duty continuously for over 19 hours. It was not until some 15 hours after the first grounding that a relief pilot was assigned to the vessel and 24 hours had elapsed on this assignment before the first pilot was relieved. The information gathered to date indicates that the pilot's fatigue probably contributed to a degradation in his ability to perform monitoring and decision-making tasks. The issue of pilot fatigue, therefore, is not limited to a single region but is widespread throughout Canada. By nature, the work environment of pilots requires them to work irregular schedules that are sometimes demanding and involve work in adverse weather conditions. In addition, pilotage regions in Canada have peak periods, when the demand for pilotage services is substantially higher than average. During such peak periods, an inadequate supply of pilots on duty, and/or work-rest schedules, may result in work-related stress and fatigue in some pilots. Sleep loss and sleepiness resulting from extended duty or altered work/rest schedules have been identified as contributory factors in many industrial accidents.[37] Studies have shown that a sleep-deprived individual tends to be a poor judge of his/her level of fatigue. Researchers at the Defence and Civil Institute of Environmental Medicine found that after 18 hours awake, people showed a 30 per cent decrement in performance on cognitive and vigilance tasks and after 48 hours the impairment averaged 60 per cent.[38] The Board believes that a flexible pilot work schedule that ensures sufficient pilots are on duty to meet the workload requirement at all times, particularly during peak periods, could alleviate performance degradation due to fatigue. Given the vulnerability of individuals in safety-sensitive positions to significant error in judgement when fatigued, and given the potential consequences of such errors, the Board recommends that: Canadian pilotage authorities adopt pilotage assignment policies and practices that both reflect the workload associated with the seasonal fluctuation in demand for pilotage services and help ensure pilots are well rested between assignments, so as to minimize the adverse effects of short-term and/or chronic fatigue on their performance. There are several ways to mitigate the negative effects of shift work and irregular work schedules. Some of the factors that affect pilot performance while on duty are controlled by the pilots themselves, in terms of modifications to their personal lifestyle, such as a combination of sleep schedules and control of one's environment and diet. Training and educational programs to help workers maintain optimum performance despite irregular work/rest schedules have been developed and can be customized to meet specific needs. The Board believes that such a fatigue management training program will assist pilots in coping with work-related stresses and fatigue. Therefore, the Board recommends that: The Department of Transport and the Canadian pilotage authorities develop and implement an awareness program to provide guidance to operational employees, including pilots, on reducing the adverse effects of fatigue on job performance. 4.2.2 BRM Training Validation On the issue of BRM training, the Board commends the action by TC and all four Canadian pilotage authorities in proposing amendments to the General Pilotage Regulations to include a requirement for pilots to be trained in BRM. The BRM training program in Canada was jointly developed by industry and the Canadian maritime educational institutions and has now been approved by TC. The objective of the training is to facilitate the effective utilization of all available resources, the exchange of vital information, and the decision-making process, through enhanced crew-pilot interaction. However, to achieve these objectives, training precepts must be effectively put into practice. In this occurrence, although the pilot had received BRM training, the elements of BRM--such as effective communication and distribution of workload based on individual expertise--which are critical to the safe operation of vessels, were not put into practice. This is not an isolated incident; at least one other recent occurrence under investigation has revealed a similar deficiency. Indeed, these BRM-related failures are found in many occurrences in the pilotage waters of Canada. Navigation with a pilot on board is a unique situation where a complete stranger, the pilot, is added to an often diverse mix of crew and they are required to function as a homogeneous team under stressful conditions and often in complex operational situations. The passive approach taken by some ships' complements and the reluctance to use the ship's complement on the part of pilots are not conducive to creating a climate whereby bridge team members can comfortably provide input to the pilot. As a result, the reluctance of pilots to include the ship's complement in the navigation process has become a norm in the pilots' culture. Consequently, pilot decision-making can become a weak link in a system prone to single-point failure. BRM training is intended to address these deficiencies. Research done into the success or failure of crew resource management training emphasizes the need for training validation to determine the effectiveness of training objectives.[39] The Board is encouraged that Canadian pilotage authorities and the majority of pilots are endorsing BRM training. However, the Board believes that BRM-related failures and deficiencies will persist if the training's precepts are not effectively transferred from the classroom to the operation front. This requires a change in attitude, culture and job behaviour. The Board believes that recurrent training, coupled with an appropriate regime to help ensure the application of learned skills and knowledge to reinforce their practice, can bring about the desired changes in existing pilot culture and behaviour. The Board, therefore, recommends that: The Department of Transport and the Canadian pilotage authorities develop and implement a Bridge Resource Management (BRM) training validation system to help ensure the principles of BRM are being put into operational practice. 4.2.3 Safety Management of Pilotage Operations Over the years, the TSB has investigated marine occurrences involving vessels operating in Canadian pilotage waters. Also, the unique nature of the master/pilot relationship was the subject of the TSB special study entitled: A Safety Study of the Operational Relationship Between Ship Masters/Watchkeeping Officers and Marine Pilots(TSB Report No. SM9501). TSB investigations revealed safety deficiencies for which recommendations were made. The safety issues that the recommendations address identify factors that affect pilot performance. These include the following: pilot/master information exchange, pilot performance degradation due to fatigue, pilot skill upgrading, training and training validation, pilot bridge resource management training and practices, and pilot fitness for duty. Investigations have also revealed that a systematic approach to fatigue management and periodic audits to evaluate pilot proficiency and skills are usually absent from pilotage organization regimes.[40] The IMO is currently identifying training needs and reviewing pilot operational procedures, aimed at setting a global common standard for pilotage expertise. In addition, at least one classification society, in conjunction with the tanker owners' organization INTERTANKO, pilotage organizations and the Norwegian coastal directorate, has introduced a certification program for the operation of pilotage organizations. The program addresses operations and management factors vital to safety and quality, including passage planning, pilot/master information exchange and definition of quality levels for such matters as training and qualifications of pilots. The Board believes that such a system will go a long way to ensuring that a pilot assigned to a vessel is fit for duty and has the necessary skills and knowledge to complete the assignment safely. Safe operation of a vessel in pilotage waters is a joint responsibility of the pilot and the master of the vessel. In recent years the IMO, recognizing the severity of the consequences of marine accidents, has been continuously striving to achieve and maintain higher standards of safety for the operation of vessels. The adoption of the mandatory International Safety Management(ISM) Code by shipowners is one of the many initiatives taken towards meeting that objective. The Board believes that as shipowners have made a commitment to quality control (through the adoption of the ISM Code), safe operation of a vessel in compulsory pilotage waters can be enhanced through a parallel, well-documented safety management/quality assurance programme adopted by pilotage authorities. The Board, therefore, recommends that: The pilotage authorities develop and implement a safety management quality assurance system, consistent with maintaining the highest practicable standard of safety, for vessels operating in Canadian pilotage waters.